Addressing missing and outlier data are common challenges in research. It’s not unusual to encounter incomplete information or extreme values in the data we receive. To ensure the reliability of our analysis, we often need to take steps to address these issues.
Missing Data
When it comes to missing data, researchers typically try to fill in the gaps using methods like estimation or prediction based on the available information. Both involve filling in the gap using inferential statistics to predict what the value likely would have been, had it not been missing. This helps ensure that we have a complete picture and can draw accurate conclusions from the data.
An example of when missing data might be filled is when a survey participant failed to answer a question on a survey. In most cases, the participant would have to be excluded for analyses involving the missing question. However, when a participant’s likely response can be estimated with high confidence, that data can be filled in.
